The Influence of the Deep Learning Approach on Students' Physics Learning Outcomes

This study aims to analyze the effect of a Deep Learning approach combined with Problem-Based Learning (PBL) on students' physics learning outcomes regarding the topic of the First Law of Thermodynamics. The urgency of this research is based on students' low understanding of the concepts of energy, heat, and work, as well as the high rate of misconceptions that impact physics learning achievement. The research method used a quantitative approach with a quasi-experimental design of the Nonequivalent Control Group Design type. The sample was determined through purposive sampling, consisting of 31 students from class XI H as the experimental group and 28 students from class XI I as the control group at MAN 2 Palu. The research instruments were a learning outcome test (pretest-posttest) and an observation sheet. Data analysis included normality tests, homogeneity tests, hypothesis testing with the Mann-Whitney U Test, and N-Gain calculation. The results showed a significant difference between the experimental and control groups (p = 0.000 < 0.05). The average N-Gain score for the experimental group reached 87.70 (high category), while the control group only reached 69.08 (medium category). Observations of the learning implementation also obtained an average score of 93% (very good category). It can be concluded that the application of the PBL-based Deep Learning approach is effective in improving the understanding of physics concepts, specifically on the material of the First Law of Thermodynamics. This approach is able to foster learning that is more mindful, meaningful, and joyful, making it relevant for supporting the achievement of 21st-century competencies.